Data Migration From SQL to MongoDB

Author Name(s): Nuzhat F. Shaikh, Aditya Jadhav, Chetan Raina, Gaurav Nagoshe, Suraj Kale
Author Email:


As the Internet related technologies have emerged, a lot of changes were observed in the data collected, stored and processed by the organizations. The new incoming data was unstructured, available from a number of outside areas and accumulated in large volumes within a small time span. It was realized the traditional system is inefficient to handle this data. Solution to that is NoSQL type database. The databases that belong to this category are highly scalable, powerful and efficient. But the problem that occurs with these databases is they don’t use the traditional SQL (Structured Query language) to query the data. Hence it is difficult to migrate data from existing traditional databases to the NoSQL databases and develop efficient and optimized functions to query the data under these NoSQL databases. Various tools are available which have the capabilities to address this, but none of them actually provide a complete integrated system for efficiently doing this. This paper proposes an architecture which will provide a GUI Application that will provide the users with a click and use Querying feature with Data Migration and Automatic Query Conversion. MongoDB is the most slanting database for NoSQL, therefore our proposed system focuses on these perspectives with reference to MongoDB.


The recent increase in the use of the internet and social media sites has emerged as a problem for the relational databases. It has been observed that the relational databases have not been able to scale up with the increased use of databases. The various factors involved such as cost and performance help in determining how the data should be stored and managed. The traditional SQL based databases are preferred for their availability and fault tolerance. But these can be very expensive to maintain and difficult to scale. This has led to corporations migrating from relational databases to NoSQL databases. Cheap hardware and easy scalability are the factors that have favored the migration to these databases. Most of the data on social networking sites and clouds is
unstructured and may vary in nature. This data also takes up huge space and requires high availability and scalability. These same tasks can still be accomplished with the traditional RDBMS databases but at a very high cost which makes the NoSQL databases the popular choice as they are comparatively cheaper [10]. MongoDB is a Document based database of NoSQL type that provides us with increased availability and scaling as its major features. It stores the data in JSON format and provides JavaScript functions to query that data. The syntax as well as the data layout being completely different, it introduces the overhead of acquiring efficiency in JavaScript querying and converting queries from all web programs from SQL to MongoDB. The proposed system focuses on 2 major aspects of the database world. Writing


The demand of NoSQL databases is increasing rapidly because of their rapid scalability and distributed architecture. Now we have a GUI that allows to implement SQL queries on migrated databases. SQL to MongoDB query conversion is now possible and getting of data from MongoDB database through SQL query is possible. The methods through which relational databases and NoSQL databases manage their information are entirely different. In RDBMS, the schema is fixed and data is stored in the form of tables where relationships may exist between multiple tables, while NoSQL databases do not follow a definite schema. In NoSQL, data is stored in an unstructured form and in a varied format of databases that includes documents, key-value pairs, columns and graphs [9]. Recently a lot of enterprises have started to migrate from SQL to NoSQL databases. The main aim of our system is to provide an environment to these enterprises to migrate data from relational database to the NoSQL data store. The data migration allows enterprise’s Online Analytical Processing (OLAP) which is a significant part of the broader category of Business Intelligence [12]. The methods used by relational and NoSQL databases to store data are absolutely different and this produces a challenging task for corporations to migrate between the two databases. The difference in the structure and nature of the RDBMS and NoSQL database make the migration process difficult. From the different choices available we have selected migration from MySQL in the SQL group to MongoDB in the NoSQL group as our test case.

795 total views, 1 views today

Download PDF File

About the author: admin